This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH 1/2] Fix incorrectly named __kernel_get_tbfreq VDSO function
On Thu, Nov 29, 2012 at 6:41 PM, Ryan S. Arnold <ryan.arnold@gmail.com> wrote:
> On Thu, Nov 29, 2012 at 6:35 PM, David Miller <davem@davemloft.net> wrote:
>> From: "Ryan S. Arnold" <ryan.arnold@gmail.com>
>> Date: Thu, 29 Nov 2012 15:07:30 -0600
>>
>>> This patch slipped through my 2.17 pre-freeze roundup for PowerPC but
>>> it's a correctness issue.
>>>
>>> Without it Tulio Quites Machado's __ppc_get_timebase_freq function
>>> ends up using the proc file system fallback for querying the timebase
>>> frequency rather than the VSYSCALL, since the name doesn't match.
>>>
>>> I'd like to check this in for 2.17 (after verification of course).
>>
>> This is fine, please commit it.
> There's something goofy happening during testing, so I'm going to see
> if I can get to the bottom of it by tomorrow.
Anton,
When this patch is applied, calls to __ppc_get_timebase_freq, which
call the libc internal __getclock_freq, return 0Hz. When the patch
isn't applied, the VSYSCALL isn't made (because the symbol match was
incorrect) and we use the proc file system fallback instead. Do you
have any idea why the kernel might be returning 0Hz with the corrected
patch?
The kernel in my test environment is 2.6.32-279.el6.ppc64 in RHEL 6.3.
I've verified that there is no difference between the 2.6 kernel code
for __kernel_get_tbfreq and that of the 3.7 kernel.
Ryan